DPMTA parameters

These parameters control the options to DPMTA, the algorithm used to provide full electrostatic interactions. DPMTA is a modified version of the FMA (Fast Multipole Algorithm) and, unfortunately, most of the parameters still refer to FMA rather than DPMTA for historical reasons. Don't be confused!

For a further description of how exactly full electrostatics are incorporated into NAMD, see section 4.5.4. For a greater level of detail about DPMTA and the specific meaning of its options, see the DPMTA distribution which is available via anonymous FTP from the site ftp.ee.duke.edu in the directory /pub/SciComp/src.

• FMA $\langle$ use full electrostatics? $\rangle$
  Acceptable Values: on or off
  Default Value: off
  Description: Specifies whether or not the DPMTA algorithm from Duke University should be used to compute the full electrostatic interactions. If set to on, DPMTA will be used with a multiple timestep integration scheme to provide full electrostatic interactions as detailed in section 4.5.4. Because of the way this feature is currently implemented, PVM must be running for DPMTA to function, even if NAMD is running only on one processor.

• FMALevels $\langle$ number of levels to use in multipole expansion $\rangle$
  Acceptable Values: positive integer
  Default Value: 5
  Description: Number of levels to use for the multipole expansion. This parameter is only used if FMA is set to on. A value of 4 should be sufficient for systems with less than 10,000 atoms. A value of 5 or greater should be used for larger systems.

• FMAMp $\langle$ number of multipole terms to use for FMA $\rangle$
  Acceptable Values: positive integer
  Default Value: 8
  Description: Number of terms to use in the multipole expansion. This parameter is only used if FMA is set to on. If the FMAFFT is set to on, then this value must be a multiple of 4. The default value of 8 should be suitable for most applications.

• FMAFFT $\langle$ use DPMTA FFT enhancement? $\rangle$
  Acceptable Values: on or off
  Default Value: on
  Description: Specifies whether or not the DPMTA code should use the FFT enhancement feature. This parameter is only used if FMA is set to on. If FMAFFT is set to on, the value of FMAMp must be set to a multiple of 4. This feature offers substantial benefits only for values of FMAMp of 8 or greater. This feature will substantially increase the amount of memory used by DPMTA.

• FMAtheta $\langle$ DPMTA theta parameter (radians) $\rangle$
  Acceptable Values: decimal
  Default Value: 0.715
  Description: This parameter specifies the value of the theta parameter used in the DPMTA calculation. The default value is based on recommendations by the developers of the code.

• FMAFFTBlock $\langle$ blocking factor for FMA FFT $\rangle$
  Acceptable Values: positive integer
  Default Value: 4
  Description: The blocking factor for the FFT enhancement to DPMTA. This parameter is only used if both FMA and FMAFFT are set to on. The default value of 4 should be suitable for most applications.

• FullDirect $\langle$ calculate full electrostatics directly? $\rangle$
  Acceptable Values: yes or no
  Default Value: no
  Description: Specifies whether or not direct computation of full electrostatics should be performed. The direct computation of electrostatics is not intended to be used during real calculations, but rather as a testing or comparison measure. Because of the $\mathcal{O}(N^2)$ computational complexity for performing direct calculations, this is much slower than using FMA to compute full electrostatics.